<snip> > We helped kill a z/OS to Windows conversion meeting with some simple > questions. > > Q. "This batch process takes 4 hours to run on z/OS. How long would you > estimate your converted process would take?" > A. "About the same amount of time." > Q. "Suppose the process terminates abnormally for some reason after about 2 > hours. We get a dump to address the problem and generally fix it from there. > Of course, we then need the full 4 hours to run the fixed program. How would > this be done on Windows?" > A. "Recompile the program with debugging enabled and then single step the > program from the command line until the problem occurs." > Q. "Are you kidding? Single step the program with the programmer monitoring > it?" > A. "Yes." > --- everybody faints in shock about how long that would take. And how much > caffeine and pizza would be consumed. > Vendor when everybody revives: "That is why once the conversion is completed > and you are running the mainframe apps on Windows, you need to do a business > process analysis to change how these processes are done. But that is not > covered in this contract." > --- some managers have another fainting spell. > > That is basically what happened. I don't know how Windows people do debugging > of an aborted production program. The above does match how I used Turbo > Pascal back when I did some minor Windows programming (Win95 time frame). > <un-snip>
http://msdn.microsoft.com/en-us/library/ms680369(v=vs.85).aspx this is really a developer training issue on windows. Windows itself support (for a long time now) both mini-dumps (z/OS problem state type information) and full crash dump (operating system memory areas and potentially cross memory, memory areas). However, if you ask, most Windows developers will not know about it. It seems to be a platform culture issue. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
